RDMA
E268844
RDMA (Remote Direct Memory Access) is a high-performance networking technology that allows one computer to directly access the memory of another without involving the operating system or CPU, significantly reducing latency and CPU overhead in data transfers.
All labels observed (4)
| Label | Occurrences |
|---|---|
| RDMA canonical | 2 |
| InfiniBand RDMA | 1 |
| RDMA (Remote Direct Memory Access) | 1 |
| RDMAP (RDMA Protocol) | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T2447786 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
Target entity: RDMA Context triple: [NVMe over Fabrics, transportIncludes, RDMA]
-
A.
InfiniBand
InfiniBand is a high-speed, low-latency communications standard commonly used as the interconnect fabric in supercomputers and large-scale data centers.
-
B.
NVMe over Fabrics
NVMe over Fabrics is a network protocol that extends the high-performance NVMe storage interface across various network transports to enable low-latency, scalable access to remote solid-state storage.
-
C.
Fibre Channel
Fibre Channel is a high-speed network technology primarily used to connect computer data storage in storage area networks (SANs), offering low-latency, high-throughput communication for enterprise environments.
-
D.
GPUDirect
GPUDirect is an NVIDIA technology that enables high-speed, low-latency data transfers directly between GPUs and other devices or memory, bypassing the CPU to improve performance in data-intensive applications.
-
E.
DMA
DMA (Direct Memory Access) is a computer feature that allows hardware devices to transfer data directly to and from system memory without continuous CPU involvement, improving performance and efficiency.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Target entity: RDMA Target entity description: RDMA (Remote Direct Memory Access) is a high-performance networking technology that allows one computer to directly access the memory of another without involving the operating system or CPU, significantly reducing latency and CPU overhead in data transfers.
-
A.
InfiniBand
InfiniBand is a high-speed, low-latency communications standard commonly used as the interconnect fabric in supercomputers and large-scale data centers.
-
B.
NVMe over Fabrics
NVMe over Fabrics is a network protocol that extends the high-performance NVMe storage interface across various network transports to enable low-latency, scalable access to remote solid-state storage.
-
C.
Fibre Channel
Fibre Channel is a high-speed network technology primarily used to connect computer data storage in storage area networks (SANs), offering low-latency, high-throughput communication for enterprise environments.
-
D.
GPUDirect
GPUDirect is an NVIDIA technology that enables high-speed, low-latency data transfers directly between GPUs and other devices or memory, bypassing the CPU to improve performance in data-intensive applications.
-
E.
DMA
DMA (Direct Memory Access) is a computer feature that allows hardware devices to transfer data directly to and from system memory without continuous CPU involvement, improving performance and efficiency.
- F. None of above. chosen
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf | computer networking technology ⓘ |
| benefit |
high throughput
ⓘ
improved scalability of distributed applications ⓘ low-latency communication ⓘ reduced CPU utilization ⓘ |
| bypasses |
remote CPU for data movement
ⓘ
remote operating system for data movement ⓘ |
| category |
high-performance network communication mechanism
ⓘ
remote memory access protocol ⓘ |
| enables | direct memory access between computers over a network ⓘ |
| fullName | Remote Direct Memory Access ⓘ |
| hasVariant |
InfiniBand
ⓘ
RoCE ⓘ RDMA over Converged Ethernet ⓘ
surface form:
RoCEv2
iWARP ⓘ |
| operatesAt | user space for data path ⓘ |
| provides |
one-sided communication primitives
ⓘ
remote read operations ⓘ remote write operations ⓘ send and receive operations ⓘ two-sided communication primitives ⓘ |
| reduces |
CPU overhead for data transfers
ⓘ
latency of data transfers ⓘ |
| requires |
RDMA-capable network fabric
ⓘ
RDMA-capable network interface controllers ⓘ memory registration ⓘ protection domains ⓘ |
| standardizedBy |
IEEE and industry consortia for RoCE
ⓘ
IETF for iWARP ⓘ InfiniBand Trade Association ⓘ |
| supports |
kernel bypass
ⓘ
zero-copy networking ⓘ |
| transportProtocolFor |
InfiniBand transport
ⓘ
RoCE over Ethernet ⓘ iWARP ⓘ
surface form:
iWARP over TCP/IP
|
| usedBy |
MPI implementations
ⓘ
NFS over RDMA ⓘ NVMe over Fabrics ⓘ SMB ⓘ
surface form:
SMB Direct
distributed key-value stores ⓘ high-performance storage systems ⓘ |
| usedIn |
data center networks
ⓘ
distributed databases ⓘ high-performance computing ⓘ message passing interfaces ⓘ storage networks ⓘ |
| uses |
completion queues
ⓘ
queue pairs ⓘ work queue elements ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Subject: RDMA Description of subject: RDMA (Remote Direct Memory Access) is a high-performance networking technology that allows one computer to directly access the memory of another without involving the operating system or CPU, significantly reducing latency and CPU overhead in data transfers.
Referenced by (5)
Full triples — surface form annotated when it differs from this entity's canonical label.